Native sim support by lgirdwood · Pull Request #10621 · thesofproject/sof

lgirdwood · 2026-03-12T20:39:19Z

Add support for native sim target and include being able to run under valgrind. This should support all cmocka tests as ztests meaning more/all can be removed. Will be added to CI soon.

Copilot

Pull request overview

Adds Zephyr native_sim support to enable running SOF Zephyr tests natively (including under Valgrind) and refines POSIX/libFuzzer integration so fuzz-specific code is only built/used when enabled.

Changes:

Add a native_sim platform target to the Zephyr build helper.
Make POSIX fuzzing sources/IPC hooks conditional on CONFIG_ARCH_POSIX_LIBFUZZER.
Extend the run scripts to support native_sim execution and optional --valgrind.

Reviewed changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
`zephyr/CMakeLists.txt`	Separates POSIX sources from fuzz-only sources so `fuzz.c` only builds when libFuzzer is enabled.
`src/platform/posix/ipc.c`	Gates fuzz ISR/IRQ plumbing behind `CONFIG_ARCH_POSIX_LIBFUZZER`.
`scripts/xtensa-build-zephyr.py`	Adds `native_sim` as a supported Zephyr platform config.
`scripts/sof-qemu-run.sh`	Adds `--valgrind` flag parsing and changes default build dir behavior for native_sim runs.
`scripts/sof-qemu-run.py`	Detects native_sim from `CMakeCache.txt`, supports running under Valgrind, and skips QEMU monitor steps for native_sim.
`app/boards/native_sim.conf`	Introduces a board-specific Kconfig fragment for native_sim.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

You can also share your feedback on Copilot code review. Take the survey.

Copilot · 2026-03-12T20:45:57Z

+#ifdef CONFIG_ARCH_POSIX_LIBFUZZER
 // Not an ISR, called from the native_posix fuzz interrupt.  Left
 // alone for general hygiene.  This is how a IPC interrupt would look
 // if we had one.


Inside the #ifdef CONFIG_ARCH_POSIX_LIBFUZZER block, the extern declaration for posix_fuzz_buf/posix_fuzz_sz (currently extern uint8_t *posix_fuzz_buf, posix_fuzz_sz;) does not match the definitions in src/platform/posix/fuzz.c (const uint8_t *posix_fuzz_buf; size_t posix_fuzz_sz;). This mismatch can cause incorrect reads/writes (e.g., posix_fuzz_sz = 0; only updating 1 byte). Please split these into separate externs with the correct types (and const).

under investigation why we dont use header to align.

tmleman · 2026-04-07T10:42:49Z

+
+if(CONFIG_SOF_BOOT_TEST_STANDALONE AND CONFIG_ZTEST)
+  set(MATH_ZTEST_SOURCES
+    ../../test/ztest/unit/math/basic/arithmetic/test_crc32_ztest.c
+    ../../test/ztest/unit/math/basic/arithmetic/test_find_equal_int16_ztest.c
+    ../../test/ztest/unit/math/basic/arithmetic/test_find_max_abs_int32_ztest.c
+    ../../test/ztest/unit/math/basic/arithmetic/test_find_min_int16_ztest.c
+    ../../test/ztest/unit/math/basic/arithmetic/test_gcd_ztest.c
+    ../../test/ztest/unit/math/basic/arithmetic/test_norm_int32_ztest.c
+  )
+
+  zephyr_library_sources(${MATH_ZTEST_SOURCES})
+
+  set_source_files_properties(${MATH_ZTEST_SOURCES}
+    PROPERTIES COMPILE_DEFINITIONS "CONFIG_NUMBERS_VECTOR_FIND=1;CONFIG_NUMBERS_NORM=1;UNIT_TEST=1"
+  )
+endif()


Either the commit message is wrong or there was some mistake here: "Make the maths tests available on native sim target.". These tests are currently built exclusively for native_sim.

tmleman · 2026-04-07T10:53:43Z

This is a good change but I would like to have the option to build native_sim in two variants. With Zephyr heap and with system heap. Please add a Kconfig enabling such a choice.

fixed, have 2 options now.

lyakh

one critical suspected breakage, should be checked at least

lyakh · 2026-05-08T06:05:02Z

+            print("\n[sof-qemu-run] No crash detected. (Skipping QEMU monitor interaction for native_sim)")
        else:
-            print("\n[sof-qemu-run] Process is no longer alive, cannot extract registers.")
+            print("\n[sof-qemu-run] No crash detected. Interacting with QEMU Monitor to grab registers...")


can the second sentence be rephrased a bit like "Requesting registers from the QEMU Monitor?" I'm not sure I'd say "I'm going to interface with my neighbour to ask if they can borrow me 2 potatoes" :-)

this reads fine.

lyakh · 2026-05-08T06:17:08Z

 	void *ptr;
 	void *raw;

+	if (bytes > 16 * 1024 * 1024) {


rmalloc_align() is also the entry point to DRAM (L3) allocations, where we already now have possibly more than 16M and we do large allocations there e.g. when downloading LLEXTs. I could imaging a very large LLEXT library with large "cold" parts that never get copied to SRAM. Can we make this something ridiculously large like 1G or maybe configurable? Not critical for now.

this caught me out by surprise as one of the tests was to intentionally alloc a large region and expect failure (to pass), but in the case of running on host we fail this test since 16M allocations are easily supported on the host. So I think today this value is fine.

lyakh · 2026-05-08T06:32:04Z

 	sof_app_main();
-#if CONFIG_SOF_BOOT_TEST && defined(QEMU_BOOT_TESTS)
+#if CONFIG_SOF_BOOT_TEST
 	sof_run_boot_tests();


I think this breaks platforms where CONFIG_BOOT_TEST_ALLOWED=y (e.g. PTL) because on them the boot test should be run later, as it's currently done.

Ack, this needs a fix.

good, this should fix the breakage, but I'm not sure about "standalone" - @kv2019i is it needed there? I assume standalone testing is working already now without this change, so wouldn't this change break them by trying to run boot tests twice?

I can drop this part for the moment, the critical part was that tests would not run under simulator without a kconfig force - this could be sim related or it could be documentation related.

@lgirdwood Took a while to check this, but I think this is ok. This change the standalone logic so that normal boot tests get also run when CONFIG_SOF_BOOT_TEST_STANDALONE is defined, but I think this is ok and makes actually sense. We are still missing some convenient method how to enable/disable certain tests when building SOF. But this is for other PRs to address.

kv2019i

Looks good, only issue is the boot test change in the second-to-last commit.

kv2019i · 2026-05-11T10:41:08Z

 	sof_app_main();
-#if CONFIG_SOF_BOOT_TEST && defined(QEMU_BOOT_TESTS)
+#if CONFIG_SOF_BOOT_TEST
 	sof_run_boot_tests();


Ack, this needs a fix.

breakage eliminated

kv2019i

My concerns addressed now.

lyakh · 2026-05-29T09:47:59Z

+                if line.startswith("CACHED_BOARD:STRING=") or line.startswith("BOARD:STRING="):
+                    if "native_sim" in line.split("=", 1)[1].strip():
+                        is_native_sim = True
+                        break


maybe using the same method as xtensa-build-zephyr.py via cmake_cache = zcmake.CMakeCache.from_build_dir(abs_build_dir.parent) would be more elegant for cmake cache parsing

Add native_sim board configuration and support in the build script. This allows building and running tests on the host using Zephyr's native_sim target. native_sim leverages the POSIX architecture, but the libfuzzer support specifically requires CONFIG_ARCH_POSIX_LIBFUZZER to be set. Therefore, this wraps fuzzer-specific code in ipc.c and the build of fuzz.c behind this config to allow clean compilation on the standard native_sim board. Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>

lyakh · 2026-06-23T07:11:00Z

+#endif
+
+
+#if defined(CONFIG_SOF_NATIVE_SIM_HOST_HEAP)


#else would do too.

Agreed in principle. This one is tied to the "move the block to the bottom" suggestion below: a bare #else here would collide with the existing #else at the non-native rmalloc section, so it only works once the native_sim block is relocated so all the non-native code is contiguous. I have addressed the functional review points in this revision; happy to do that structural consolidation (single #if !defined(...) ... #else ... #endif with native_sim as the trailing #else) as a focused follow-up so this diff stays reviewable — let me know if you'd prefer it folded in now.

lyakh · 2026-06-23T07:11:39Z

+
+	return ptr;
+}
+EXPORT_SYMBOL(rmalloc_align);


I don't think you need LLEXT symbol exporting for the simulator

Done — dropped the EXPORT_SYMBOL() calls from the native_sim allocator path (rmalloc_align/rmalloc/rzalloc/rballoc_align/rfree); the simulator does not load LLEXT modules so it does not need them.

lyakh · 2026-06-23T07:12:41Z

+	rfree(addr);
+}
+
+#else


maybe put this whole block at the bottom after line 777 (original 674)

Makes sense — consolidating the native_sim implementation into one trailing block (the #else of the !defined(NATIVE_SIM) section) would remove the fragmentation. As noted on the #else comment above, I kept the current structure for this revision to keep the diff focused on the functional fixes; I can do the relocation as a follow-up commit if you want it in this PR.

Copilot

Pull request overview

Copilot reviewed 9 out of 9 changed files in this pull request and generated 5 comments.

lgirdwood · 2026-06-24T13:59:31Z

+	if (alignment < sizeof(void *))
+		alignment = sizeof(void *);
+
+	raw = nsi_host_malloc(bytes + alignment + sizeof(void *));
+	if (!raw)
+		return NULL;
+
+	ptr = (void *)(((uintptr_t)raw + sizeof(void *) + alignment - 1) & ~(alignment - 1));
+	((void **)ptr)[-1] = raw;


Fixed the power-of-two assumption: alignment is now rounded up to the next power of two before the mask-based round-up, so a non-power-of-two value can no longer yield a misaligned pointer. On the overflow point: this is the native_sim host build (64-bit size_t) and bytes is already capped at 16MB just above, so bytes + alignment + sizeof(void *) cannot overflow; I added a comment noting that rather than a dead runtime check.

lgirdwood · 2026-06-24T13:59:33Z

+void *sof_heap_alloc(struct k_heap *heap, uint32_t flags, size_t bytes,
+		     size_t alignment)
+{
+	return rmalloc_align(flags, bytes, alignment);
+}
+
+void sof_heap_free(struct k_heap *heap, void *addr)
+{
+	rfree(addr);
+}


Fixed — the native_sim path now defines z_impl_sof_heap_alloc()/z_impl_sof_heap_free() and only defines the plain sof_heap_alloc()/sof_heap_free() wrappers under #ifndef CONFIG_SOF_USERSPACE_INTERFACE_ALLOC, mirroring the non-native path. So with the syscall interface enabled the generated wrappers bind to z_impl_*, and there is no longer a conflicting direct definition.

lgirdwood · 2026-06-24T13:59:27Z

    # Make absolute path just in case
+    # The shell script cd's into `args.build_dir` before executing us, so `args.build_dir` might be relative to the shell script's pwd.
+    # We resolve it relative to the python script's original invocation cwd.
    build_dir = os.path.abspath(args.build_dir)


Fixed — updated the comment; the shell wrapper no longer cd's, so it now just notes that args.build_dir may be relative to the caller's cwd and is resolved to absolute.

lgirdwood · 2026-06-24T13:59:29Z

@@ -1195,7 +1199,7 @@ def install_platform(platform, sof_output_dir, platf_build_environ, platform_wco

 	os.makedirs(install_key_dir, exist_ok=True)
 	# looses file owner and group - file is commonly accessible, dont install qemu.


Fixed — "looses" -> "loses", "dont" -> "don't".

lgirdwood · 2026-06-24T13:59:25Z

+#if CONFIG_SOF_BOOT_TEST && (defined(QEMU_BOOT_TESTS) || CONFIG_SOF_BOOT_TEST_STANDALONE)
 	sof_run_boot_tests();
 	qemu_xtensa_exit(0);
 #endif


Fixed — the qemu_xtensa_exit(0) call is now wrapped in #if defined(QEMU_BOOT_TESTS), so a standalone boot test (e.g. native_sim) just returns from test_main() instead of referencing the QEMU-only symbol.

Add native_sim board target to the sof-qemu-run scripts, and add an option to additionally run it under valgrind. The default build directory is set to ../build-native_sim Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>

When building the firmware for native_sim, debugging allocations with host machine tools like Valgrind is constrained due to Zephyr's internal minimal libc tracking the heap manually via static pools. By bypassing Zephyr's memory interception on native_sim using nsi_host_malloc, dynamically tracked memory can surface appropriately to Valgrind memory checkers without causing a libc heap pool panic. Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>

Keep spinning in case user needs to inspect status via monitor. Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>

When building the native_sim fuzzer, the host allocator does not possess the strict bounds of the internal Zephyr memory pools. If the fuzzer generates a malformed payload requesting an excessively large size (e.g. 4GB), it passes directly to the host ASAN allocator which aborts due to OOM or protection limits. Adding a 16MB cap allows these to fail gracefully. Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>

Only run sof_run_boot_tests() at boot for QEMU or native_sim (STANDALONE) targets. Other platforms trigger via IPC.

lgirdwood requested review from abonislawski, dbaluta, iuliana-prodan, kv2019i, lbetlej, lyakh, mmaka1 and plbossart as code owners March 12, 2026 20:39

Copilot AI review requested due to automatic review settings March 12, 2026 20:39

Copilot started reviewing on behalf of lgirdwood March 12, 2026 20:41 View session

Copilot AI reviewed Mar 12, 2026

View reviewed changes

lyakh reviewed Mar 13, 2026

View reviewed changes

Comment thread scripts/sof-qemu-run.py Outdated

Comment thread scripts/sof-qemu-run.sh Outdated

lgirdwood force-pushed the native_sim branch from 68f8b9a to 6ba3347 Compare March 27, 2026 16:33

lgirdwood requested review from dabekjakub and singalsu as code owners March 27, 2026 16:33

lgirdwood force-pushed the native_sim branch from 6ba3347 to 9c5a577 Compare March 27, 2026 18:59

lgirdwood requested review from bardliao, pblaszko and tmleman as code owners March 27, 2026 21:27

lgirdwood marked this pull request as draft March 27, 2026 21:33

lgirdwood force-pushed the native_sim branch 2 times, most recently from 11e03f6 to db6ddb6 Compare March 31, 2026 09:18

lgirdwood marked this pull request as ready for review March 31, 2026 09:39

lgirdwood force-pushed the native_sim branch from db6ddb6 to a6e0a25 Compare March 31, 2026 11:43

tmleman reviewed Apr 7, 2026

View reviewed changes

tmleman requested changes Apr 7, 2026

View reviewed changes

lgirdwood force-pushed the native_sim branch 3 times, most recently from f2a31fa to 4dbff45 Compare May 7, 2026 13:08

lyakh previously requested changes May 8, 2026

View reviewed changes

kv2019i requested changes May 11, 2026

View reviewed changes

kv2019i mentioned this pull request May 11, 2026

[SKIP SOF-TEST] .github:: add pull request workflow for Zephyr shell builds #10763

Merged

lgirdwood force-pushed the native_sim branch 2 times, most recently from 37323c5 to bb6566e Compare May 26, 2026 20:49

kv2019i approved these changes May 29, 2026

View reviewed changes

lyakh reviewed May 29, 2026

View reviewed changes

lgirdwood force-pushed the native_sim branch 2 times, most recently from 3ed3179 to 67b92fd Compare June 19, 2026 12:08

lyakh reviewed Jun 23, 2026

View reviewed changes

lgirdwood requested a review from Copilot June 23, 2026 11:18

Copilot started reviewing on behalf of lgirdwood June 23, 2026 11:18 View session

Copilot AI reviewed Jun 23, 2026

View reviewed changes

lrgirdwo added 5 commits June 24, 2026 14:58

qemu: exit for boot test only otherwise spin

614fc04

Keep spinning in case user needs to inspect status via monitor. Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>

zephyr: native_sim: run boot tests at boot if STANDALONE enabled

03ebb6b

Only run sof_run_boot_tests() at boot for QEMU or native_sim (STANDALONE) targets. Other platforms trigger via IPC.

lgirdwood force-pushed the native_sim branch from 67b92fd to 03ebb6b Compare June 24, 2026 13:58

		@@ -1195,7 +1199,7 @@ def install_platform(platform, sof_output_dir, platf_build_environ, platform_wco

		os.makedirs(install_key_dir, exist_ok=True)
		# looses file owner and group - file is commonly accessible, dont install qemu.

Uh oh!

Conversation

lgirdwood commented Mar 12, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Copilot AI Mar 12, 2026

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

lyakh left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

kv2019i left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

kv2019i left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Choose a reason for hiding this comment